A pythonic tool for batch loading data files (json, parquet, csv, tsv) into ElasticSearch
Project description
Main features
Batch upload CSV (actually any *SV) files to Elasticsearch
Batch upload JSON files / JSON lines to Elasticsearch
Batch upload parquet files to Elasticsearch
Pre defining custom mappings
Delete index before upload
Index documents with _id from the document itself
Load data directly from url
SSL and basic auth
Unicode Support ✌️
Plugins
In order to install plugin, simply run pip install plugin-name - esl-redis - Read continuously from a redis list(s) and index to elasticsearch - esl-s3 - Plugin for listing and indexing files from S3
Test matrix
python / es |
5.6.16 |
6.8.0 |
7.1.1 |
|---|---|---|---|
2.7 |
V |
V |
V |
3.7 |
V |
V |
V |
Installation
Usage
(venv)/tmp $ elasticsearch_loader --help
Usage: elasticsearch_loader [OPTIONS] COMMAND [ARGS]...
Options:
-c, --config-file TEXT Load default configuration file from esl.yml
--bulk-size INTEGER How many docs to collect before writing to
Elasticsearch (default 500)
--es-host TEXT Elasticsearch cluster entry point. (default
http://localhost:9200)
--verify-certs Make sure we verify SSL certificates
(default false)
--use-ssl Turn on SSL (default false)
--ca-certs TEXT Provide a path to CA certs on disk
--http-auth TEXT Provide username and password for basic auth
in the format of username:password
--index TEXT Destination index name [required]
--delete Delete index before import? (default false)
--update Merge and update existing doc instead of
overwrite
--progress Enable progress bar - NOTICE: in order to
show progress the entire input should be
collected and can consume more memory than
without progress bar
--type TEXT Docs type. TYPES WILL BE DEPRECATED IN APIS
IN ELASTICSEARCH 7, AND COMPLETELY REMOVED
IN 8. [required]
--id-field TEXT Specify field name that be used as document
id
--as-child Insert _parent, _routing field, the value is
same as _id. Note: must specify --id-field
explicitly
--with-retry Retry if ES bulk insertion failed
--index-settings-file FILENAME Specify path to json file containing index
mapping and settings, creates index if
missing
--timeout FLOAT Specify request timeout in seconds for
Elasticsearch client
--encoding TEXT Specify content encoding for input files
--keys TEXT Comma separated keys to pick from each
document
-h, --help Show this message and exit.
Commands:
csv
json FILES with the format of [{"a": "1"}, {"b": "2"}]
parquet
redis
s3
Examples
Load 2 CSV to elasticsearch
elasticsearch_loader --index incidents --type incident csv file1.csv file2.csv
Load JSONs to elasticsearch
elasticsearch_loader --index incidents --type incident json *.json
Load all git commits into elasticsearch
git log --pretty=format:'{"sha":"%H","author_name":"%aN", "author_email": "%aE","date":"%ad","message":"%f"}' | elasticsearch_loader --type git --index git json --json-lines -
Load parquet to elasticsearch
elasticsearch_loader --index incidents --type incident parquet file1.parquet
Load CSV from github repo (actually any http/https is ok)
elasticsearch_loader --index data --type avg_height --id-field country json https://raw.githubusercontent.com/samayo/country-data/master/src/country-avg-male-height.json
Load data from stdin
generate_data | elasticsearch_loader --index data --type incident csv -
Read id from incident_id field
elasticsearch_loader --id-field incident_id --index incidents --type incident csv file1.csv file2.csv
Load custom mappings
elasticsearch_loader --index-settings-file samples/mappings.json --index incidents --type incident csv file1.csv file2.csv
Tests and sample data
End to end and regression tests are located under test directory and can run by running ./test.py Input formats can be found under samples
